Reply to Harry Herr's Letter to the Editor re: Marko Babjuk, Andreas Bohle, Maximilian Burger, et al. EAU Guidelines on Non-muscle-invasive Urothelial Carcinoma of the Bladder: Update 2016. Eur Urol 2017;71:447-61
Abstract
The definition of failure of bacillus Calmette-Guérin (BCG) intravesical immunotherapy has been controversial for many years. As consequence, different definitions are used in the literature [1.
One of the most controversial issues is the definition of the interval after initiation of BCG treatment when we can say that BCG has failed and patients are unlikely to respond to further BCG instillations. The importance of the correct definition is driven by the danger of muscle-invasive progression, which is associated with extremely poor prognosis [2.
However, many patients are unwilling to undergo radical cystectomy without exhausting all the less aggressive options. Several analyses have demonstrated unfavourable prognosis in patients with tumour persistence immediately after an induction course of BCG.
Solsona et al [3 showed that the 3-mo response was the only independent factor predictive of invasive progression. In a SWOG study, patients who achieved a complete response at 3 mo had 5-yr survival probability of 77%, compared to 62% for patients who did not [4.
In an analysis of CUETO trials, tumour presence at 3-mo cystoscopy was associated with a higher risk of progression in multivariate analysis [5. In analysis published in 2003, Herr and Dalbagni [6 favoured 6-mo over 3-mo response in prediction of further recurrences.
Unfortunately, a similar analysis for progression was not presented, probably because of a low number of events. As tumour progression and survival, in contrast to further recurrences, are the main issue for patients with high-risk non-muscle-invasive bladder cancer (NMIBC) after BCG, the EAU guidelines panel decided to set status at 3 mo as crucial for the definition of BCG failure for high-grade papillary tumours.
We greatly appreciate the thorough analysis presented by Dr. Herr [7.
In spite of small discrepancies in numbers (789 vs 784 patients with no tumour at 3 and 6 mo), it is based on previously published data for patients with high-risk NMIBC treated with BCG without maintenance [8. The letter by Dr.
Herr contains an additional analysis of progression-free survival according to status at 3 and 6 mo after BCG initiation (Fig. 1 in [7), which could be, as soon as it is published, the key for a potential change in guideline recommendations. The analysis demonstrated that persistent disease at 3 mo after induction BCG disappeared in 23% of patients after a further 3 mo, even without additional BCG.
A more important group, however, is the 77% of patients with tumours that still persist at 6 mo, whereby the curves presented show a dramatic 60% 2-yr progression rate. How many of those patients could we render "progression-free" by performing an early radical cystectomy at 3 mo, with potentially better disease-related survival? This is why we should be very cautious in our recommendations.
The analysis raises several further questions. Patients with tumour present at 3 and 6 mo have similar progression-free survival to patients with an initial response and early recurrence at 6 mo.
However, these may represent different biological entities. We all feel that the timing of recurrence after initial response to BCG may have prognostic implications, and earlier recurrence may be associated with poorer prognosis [9, but this must be specified in further prospective trials.